voice_audio_multi/include/engine/community_models/latentsync/sampling.h
Fabio c36510c21e
Some checks failed
Linux Build / Linux cpu (push) Has been cancelled
Linux Build / Linux cpu-clang (push) Has been cancelled
Linux Build / Linux vulkan (push) Has been cancelled
mac-build / macOS CPU (push) Has been cancelled
Nix Build / Nix (cpu) (push) Has been cancelled
Nix Build / Nix (vulkan) (push) Has been cancelled
windows-build / Windows CPU (push) Has been cancelled
Initial import: audio.cpp with Orange Pi documentation
2026-11-04 19:05:22 +01:00

32 lines
972 B
C++

#pragma once
#include <cstdint>
#include <cstddef>
#include <vector>
namespace engine::models::latentsync {
struct LatentSyncDdimConfig {
int64_t training_steps = 1000;
float beta_start = 0.00085F;
float beta_end = 0.012F;
int64_t steps_offset = 1;
};
// Official epsilon-prediction DDIM schedule: scaled-linear betas, eta=0,
// no sample clipping, and final alpha equal to the first cumulative alpha.
class LatentSyncDdimSchedule {
public:
LatentSyncDdimSchedule(LatentSyncDdimConfig config, int64_t inference_steps);
const std::vector<int64_t> & timesteps() const noexcept { return timesteps_; }
void step(std::vector<float> & sample, const std::vector<float> & noise, size_t index) const;
private:
std::vector<float> alphas_;
std::vector<int64_t> timesteps_;
int64_t stride_;
};
std::vector<float> latentsync_timestep_features(int64_t timestep, int64_t channels, int64_t batch);
} // namespace engine::models::latentsync